ORIGINAL RESEARCH Protective role of bacillithiol in superoxide stress and Fe–S metabolism in Bacillus subtilis Zhong Fang & Patricia C. Dos Santos Department of Chemistry, Wake Forest University, Winston-Salem, North Carolina 27016.

Keywords Abstract Bacillithiol, dehydratase, Fe–S cluster, Suf, superoxide stress. (GSH) serves as the prime in most organisms as its depletion increases antibiotic and metal toxicity, impairs oxidative stress responses, and Correspondence affects Fe and Fe–S cluster metabolism. Many gram-positive bacteria lack GSH, Patricia C. Dos Santos, Department of but instead produce other structurally unrelated yet functionally equivalent thi- Chemistry, Wake Forest University, Winston- ols. Among those, bacillithiol (BSH) has been recently identified in several low Salem, NC 27016. Tel: 336-758-3144; Fax: G+C gram-positive bacteria. In this work, we have explored the link between 336-758-4656; E-mail: [email protected] BSH and Fe–S metabolism in Bacillus subtilis. We have identified that B. subtilis lacking BSH is more sensitive to oxidative stress (paraquat), and metal toxicity Funding Information (Cu(I) and Cd(II)), but not H2O2. Furthermore, a slow growth phenotype of This research was funded by the National BSH null strain in minimal medium was observed, which could be recovered Science Foundation (MCB-1054623). upon the addition of selected amino acids (Leu/Ile and Glu/Gln), supplementa- tion of iron, or chemical complementation with BSH disulfide (BSSB) to the Received: 12 March 2015; Revised: 16 April growth medium. Interestingly, Fe–S cluster containing isopropylmalate isomer- 2015; Accepted: 17 April 2015 ase (LeuCD) and glutamate synthase (GOGAT) showed decreased activities in BSH null strain. Deficiency of BSH also resulted in decreased levels of intracel- MicrobiologyOpen 2015; 4(4): 616–631 lular Fe accompanied by increased levels of manganese and altered expression – doi: 10.1002/mbo3.267 levels of Fe S cluster biosynthetic SUF components. Together, this study is the first to establish a link between BSH and Fe–S metabolism in B. subtilis.

Introduction in its reduced form and its disulfide form (GSSG) with the cellular ratio of GSH:GSSG estimated to be around Biothiols are involved in numerous cellular functions, 100:1 (Fahey et al. 1978; Meister and Anderson 1983) as including but not limited to a variety of biosynthetic opposed to cysteine, whose thiol:disulfide ratio lies pathways, detoxification by conjugation, metal metabo- around 25:1 (Newton et al. 2009; Sharma et al. 2013). lism, and cell division (Winterbourn and Metodiewa Early studies have shown that GSH plays a key role in a 1999; Jacob et al. 2006; Hider and Kong 2011). They can wide array of biochemical functions. First, GSH can be classified as large molecular weight , also called directly serve as an by donating electrons to protein thiols, and low molecular weight (LMW) free thi- reactive radicals through autoxidation reactions to form ols (Sen and Packer 2000). Cysteine is the most common GSSG (Meister and Anderson 1983; Meister 1988; Ches- LMW biothiol and its sulfhydryl group provides unique ney et al. 1996). Thus, under oxidative stress conditions, chemical functionality distinct from other standard amino the GSH/GSSG ratio may fall into the 1–10 range (Reed acids. Its neutral pKa, nucleophilicity, redox properties, 1990). Second, GSH serves as the substrate of biochemical metal ion affinity, and bonding characteristics make Cys a reactions. For example, dehydroascorbate reductase which versatile site for many chemical processes (Bulaj et al. is a key component of the GSH-ascorbate cycle uses GSH

1998; Jacob et al. 2006; Roos et al. 2013). as a substrate to produce ascorbate to detoxify H2O2 In the late 19th century, glutathione (GSH) was identi- (Crook 1941). Besides the various roles of GSH on detox- fied as another abundant LMW thiol widely distributed ifying oxidants and toxins, GSH is also involved in in most cells (Rey-Pailhade 1888a,b). GSH mainly exists chemical modification of protein cysteine residues

616 ª 2015 The Authors. MicrobiologyOpen published by John Wiley & Sons Ltd. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Z. Fang & P. C. Dos Santos Protective Role of Bacillithiol

(Dalle-Donne et al. 2009). S-glutathionylation is consid- exerted a small but significant protective effect ered a specific posttranslational modification on protein against WhiD cluster loss under oxidative stress (Crack cysteine residues in response to oxidative stress or nitro- et al. 2009). sative stress (Giustarini et al. 2005; Martinez-Ruiz and In some low-(G+C)-content gram-positive bacteria, Lamas 2007; Dalle-Donne et al. 2009). Several findings such as Bacillus species, bacillithiol (BSH) is a major have also reported the vital role for S-glutathionylation in LMW thiol recently discovered in bacteria lacking GSH redox-regulation, energy metabolism, cellular signaling, and MSH (Newton et al. 2009). Because of the crucial calcium homeostasis, protein folding, and stability (Reed roles of GSH and MSH in thiol-redox homeostasis, pro- 1990; Cotgreave et al. 2002; Dalle-Donne et al. 2003, tein posttranslational modification, and xenobiotic 2009; Pan and Berk 2007; Demasi et al. 2008; Rodriguez- detoxification, it is anticipated that some of these func- Pascual et al. 2008). tions might also be performed by BSH (Helmann 2011). Interestingly, the effect of GSH on the superoxide- The BSH redox potential was characterized to be sensitive Fe–S cluster-containing aconitase (ACN) of 221 mV through in vitro coupled assay with GSH/ Escherichia coli has also been demonstrated (Gardner GSSG, which is higher than Cys (223 mV) and Co- and Fridovich 1993a). The results showed a ~20% ASH (234 mV) (Sharma et al. 2013). This implies that decrease in ACN activity in a GSH-deficient E. coli BSH does not have better capacity to buffer oxidative strain, displaying a phenotype which could be recovered stress through autoxidation reaction. However, consider- upon addition of GSH to the growth medium (Gardner ing the low thiol pKa value of BSH (pKaSH/S = 7.97) and Fridovich 1993a). Earlier research indicated that and the relatively high concentration of BSH (0.5– the reactivation of O2 -inactivated ACN in cell extract 5 mmol/L) (Sharma et al. 2013), it is still possible that was blocked by adding ethylenediaminetetraacetic acid BSH is preferable to Cys and CoA as a major redox buf- (EDTA) or 2,20-dipyridyl (DP) (Kennedy et al. 1983; fer in B. subtilis. In addition, BSH has been shown to Gardner and Fridovich 1992). This observation led to modulate the intracellular labile zinc pool by forming a the proposal that the reactivation of O2 -inactivated tetrahedral BSH2:zinc complex with two sulfur and two ACN would involve the restoration of [3Fe–4S] clusters oxygen ligands (Ma et al. 2014). Although BSH is not by free Fe2+. Overall, in this model, GSH would act as an essential metabolite under standard laboratory condi- either a reductant or Fe2+ donor to facilitate repair of tions, initial studies have identified that bacterial strains damaged Fe–S clusters. Recently, it has been reported unable to synthesize this thiol displayed impaired sporu- that GSH also acts as a key component of the cytoplas- lation, sensitivity to acid and salt, depleted NADPH mic labile iron pool providing further support for the level, and increased sensitivity to , hypochlo- important role of GSH in iron homeostasis (Hider and rite, diamide, and methylglyoxal (Gaballa et al. 2010; Kong 2011). Moreover, changes in the redox ratio of Lamers et al. 2012; Roberts et al. 2013; Chandrangsu GSH/GSSG has been correlated with induction of man- et al. 2014; Posada et al. 2014). While defects in metab- ganese (Mn)-dependent superoxide dismutase expression olism resulting from BSH phenotypes have been identi- (Gardner and Fridovich 1993b). Despite compelling evi- fied, biochemical pathways involving BSH have not been dence for GSH’s involvement in oxidative stress, thiol fully explored. homeostasis, and Fe–S cluster metabolism, the complete The goal of this work was to investigate the role of inventory of biological processes involving GSH as well BSH in Fe–S metabolism under normal and stress con- as the exact mechanism of GSH-dependent reactions is ditions. We show that lack of BSH caused a slow still not fully understood. growth phenotype in minimal medium, which may be Although GSH is recognized as the dominant LMW due to the lower activity of Fe–S enzymes involved in thiol in eukaryotes and gram-negative prokaryotes, GSH branched chain amino acid biosynthesis. Furthermore, is not detected in many gram-positive species. This growth assays under various stress conditions revealed a observation suggested the existence of other LMW thiols protective role of BSH against copper stress, superoxide replacing the role of GSH in those species. In fact, in stress, cadmium stress, and iron starvation. Along with Actinobacteria, mycothiol (MSH) was identified as a the parallel Fe–S enzyme assays, thiol analysis, and novel cysteine derivative (Newton et al. 1996). Mycobac- intracellular metal analysis suggest the involvement of terium smegmatis strains lacking mycothiol are more BSH in superoxide stress and metal homeostasis. More- sensitive to reactive oxygen and nitrogen species, such as over, western blot analysis was performed on Fe–S hydrogen , menadione, plumbagin, and t-butyl cluster biosynthetic proteins in wild-type and mutant hydrogen peroxide (Newton et al. 1999, 2008; Rawat strains. Metal analysis of a BSH deletion strain under et al. 2002). A recent study on WhiD, a member of the paraquat (PQ) stress also revealed a link between BSH WhiB-like family of Fe–S proteins, indicated that methyl and Mn homeostasis.

ª 2015 The Authors. MicrobiologyOpen published by John Wiley & Sons Ltd. 617 Protective Role of Bacillithiol Z. Fang & P. C. Dos Santos

Experimental Procedures anaerobically. Clear cell extracts were collected by centri- fugation at 13,200g for 10 min and stored in sealed vials for assays. Protein concentrations were determined by Reagents Bradford protein assay using BSA as standard. 3-isopropylmalic acid and monobromobimane (mBBr) Glutamate synthase (GOGAT) assays were conducted were purchased from Sigma-Aldrich (St. Louis, MO, using the protocol as described in reference (Outten et al. USA). N-acetyl-glucosamine-malate and BSH disulfide 2004) with minor modifications. To ensure consistent reac- (BSSB) were synthesized as described in Sharma et al. tion conditions, 200 mmol/L Tris-HCl (pH 8.0), 0.1 mol/L (2011) and Lamers et al. (2012), respectively. Primary buffered a-ketoglutarate, and 8 mmol/L NADPH were antibodies were generated by Thermo Scientific Pierce mixed in a ratio of 32:2:1 to make a stock reaction mix. In (Waltham, MA, USA) using laboratory purified protein each assay, 875 lL of fresh stock reaction mix and 75 lLof antigen (SufC, SufD, or MnmA) for rabbit immunization, clear cell lysate were injected into a sealed cuvette filled whereas Goat anti-Rabbit IgG (H+L) with AP conjugate with argon gas. The decreasing slope of absorbance at as secondary antibody and Lumi-Phos WB chemilumines- 340 nm (NADPH oxidation) was recorded as S1, and then cent substrate were both from Thermo Scientific. All the 50 lL of 0.2 mol/L L-glutamine was added and second other reagents were obtained from Fisher Scientific and slope was recorded as S2. The specific activity of GOGAT Sigma-Aldrich. was calculated by the following equation:

S2 1:05 S1 SAðlmol min 1 mg 1Þ¼ Bacterial growth 0:00622 mg total protein Bacillus subtilis CU1065 (wt), HB11002 (ΔbshA), HB110079 Isopropylmalate isomerase (LeuCD) was assayed by

(ΔbshC), and HB110091 (ΔbshC amyE::PxylAbshC)strains using 3-isopropylmalic acid as substrate and by following were kindly provided by Prof. John Helmann (Cornell Uni- the formation of the intermediate dimethylcitraconate at versity) (Gaballa et al. 2010). Strains were outgrown either 235 nm (Fultz and Kemper 1981; Manikandan et al. 2011). in Luria broth (LB) medium or Spizizen’s minimal In brief, 800 lL of 200 mmol/L phosphate buffer (pH 7.0) medium plus 10 lg/mL tryptophan (MM). For the growth and 100 lL of cell extract were mixed first, followed by experiments, supplementary components were added addition of 100 lL of 10 mmol/L 3-isopropylmalic acid to where appropriate to the following concentrations: casami- start the reaction. Blanks were carried out in the absence of 1 no acid, 0.05%; L-glutamine, 120 lg/mL; L-glutamate, substrate, and extinction coefficient of 4.668 mmol L 1 120 lg/mL; L-leucine, 46 lg/mL; L-isoleucine, 34 lg/mL. cm was used to calculate the activity. When antibiotics were needed, medium was supplemented ACN catalyzes a reversible reaction between citrate and with 12.5 lg/mL erythromycin and 12.5 lg/mL lincomycin iso-citrate through an intermediate cis-aconitate which has or 40 lg/mL kanamycin. In growth assays, strains cultured absorption at 240 nm (Alen and Sonenshein 1999). Each l up to OD600 of 1.0 (measured using a 1 cm cuvette) in assay reaction contained 700 L of 100 mmol/L Tris-HCl, l l MM were diluted to OD600 of 0.05 into 200 lL of fresh pH 8.0, 50 L of cell extract and 100 L of 200 mmol/L medium and allowed to grow at 37°C on Falcon 96-well Tris-Na-Citrate. The increase in absorbance at 240 nm was plate. The cell growth rate was monitored at Abs600 using measured and a millimolar extinction coefficient of Synergy H1 plate reader from BioTek (Winooski, VT, 3.4 mmol L 1 cm 1 was used to calculate the activities. USA). Control reactions were performed without substrate. The activity of MDH was measured by following the consumption of NADH at 340 nm (Siegel 1969). The Enzymatic assays reaction mix contained 50 lmol of Tris-HCl (pH 7.4), Activities of Fe–S cluster containing proteins, mononu- 0.38 lmol of oxaloacetate, 0.15 lmol NADH and 50 lL clear iron enzymes and the non-Fe–S cluster containing of cell extract in a total volume of 1 mL. The decrease in malate dehydrogenase (MDH) were determined in clear absorption at 340 nm was measured over time. cell-free lysate. Bacillus subtilis wild-type and ΔbshA were Quercetin 2,3-dioxygenase (QD) catalyzes the cleavage grown in MM to OD600 of 0.8–1.0. Cells were harvested of the O-heteroaromatic ring of quercetin to produce 2- by centrifugation at 5000g for 10 min and frozen in protocatechuoylphloroglucinol carboxylic acid and carbon 80°C until use. Cell pellets from 500 mL cultures were monoxide. QD activity was measured based on the transferred into anaerobic glove box and washed by 5 mL decomposition of quercetin as indicated by the decrease of 25 mmol/L Tris-HCl, 150 mmol/L NaCl, 10% glycerol, in absorption at 367 nm (Bowater et al. 2004). Cell lysate pH 8.0 degassed buffer. The resulting pellets were then was prepared anaerobically as described above and by resuspended in 2 mL buffer and disrupted by sonication adding 5 lmol/L quercetin in lysis buffer to stabilize

618 ª 2015 The Authors. MicrobiologyOpen published by John Wiley & Sons Ltd. Z. Fang & P. C. Dos Santos Protective Role of Bacillithiol metal cofactor. A typical assay reaction contained 1 mL All the samples were analyzed using inductively coupled of 200 mmol/L Tris-HCl (pH, 8.0), 100 lL of crude plasma optical emission spectrometry (ICP-OES) from extract and 50 lL of 1.2 mmol/L quercetin in dimethyl Teledyne Leeman Labs (Hudson, NH, USA). The wave- sulfoxide. The absorption at 367 nm was monitored over lengths for the detection of each metal were as follows time for 2 min and an extinction coefficient of 20 mmol/ (nm): Fe, 238.204, 239.563; Mn, 257.610, 259.372; Sample L 1 cm 1 was used to calculate the activity. concentrations were calculated using metal standard solu- Assays of threonine dehydrogenase (Tdh) were per- tions and normalized to the dry weights (DWs) of cell formed by following the previous reported protocol (Gu pellets. and Imlay 2013). The cell lysate was prepared anaerobi- cally using Tris-HCl buffer plus 5 mmol/L L-threonine. In Quantification of thiol levels under stress each assay, 600 lL of 50 mmol/L Bicine (pH, 8.5), conditions 100 lL of cell extract and 20 lL of 50 mmol/L NAD were mixed in a sealed cuvette, and then 300 lL of buffered The contents of reduced thiol BSH, oxidized thiol BSSB 90 mmol/L L-threonine was added to initiate the reaction. and protein thiol (BSSProt) were prepared as described by The activity of Tdh was measured by monitoring the pro- Newton et al. (2009), Chi et al. (2013). In brief, 25 mL of duction of NADH at Abs340. cultures in MM were challenged with stress at OD600 of 1.0 and harvested at 0, 20, and 50 min after exposure. Controls were prepared by adding 5 mmol/L of N-ethylmaleimide ICP-OES analysis for cellular metal (NEM) to block-free thiol, and followed by derivatization concentration with 2.5 mmol/L of mBBr. The preparations of disulfide Bacillus subtilis wild-type and ΔbshA strains were grown in samples were similar to controls, except being reduced with

MM until OD600 reached 1.0. Culture aliquots (300 mL) dithiothreitol (DTT) prior derivatization. Reduced thiol were removed before the addition of 100 lmol/L PQ, and was prepared by direct incubation with 2 mmol/L mBBr. 20 min, 50 min after stress exposure. Cells were harvested Total thiol content was measured by incubating cell pellet by centrifugation and washed twice with 10 mL of chilled with 500 lL of 25 mmol/L 4-(2-hydroxyl)-1-piperazine- 25 mmol/L phosphate buffer (pH.7.4) containing 1 mmol/ ethanesulfonic acid (HEPES), 2 mmol/L DTT, pH 8.0 on L EDTA. The resulting cell pellets were immediately washed ice for 1 h, followed by adding 500 lL of 8 mmol/L mBBr twice with the same buffer without EDTA and resuspended in acetonitrile. The mixture was incubated at 65°C for with 10 mL of phosphate buffer. Aliquots of resuspended 20 min and quenched by the addition of 10 lLof2N cells (2 mL) were transferred to an eppendorf tube, in HCl. All samples were diluted five times with 5 mmol/L which 20 lL of 20 mg/mL lysozyme was added. The sam- HCl and injected 50 lL for high performance liquid chro- ple was incubated at 37°C for 30 min, followed by adding matography (HPLC) analysis. The HPLC method was con- 2 mL of 5% nitric acid with 0.1% (v/w) Triton X-100. The ducted with a Waters Symmetry C8 column (3.9 9 sample was digested at 95°C for 30 min and supernatant 150 mm, 5 lm) using the gradient as described in (New- was collected by centrifugation. Each supernatant was ton et al. 2009). The intracellular concentrations of BSH diluted with an equal volume of deionized water for the derivatives were calculated based on BSH standard and measurement of total metal content. normalized to cell DW. The concentrations of labile metal (defined as metals not associated with a higher molecular weight fraction, Western blot analysis <3000 Da) were quantified as follow. Eight milliliters of cell suspension from above was disrupted by an Avestin Emul- Cells were grown in MM to OD600 of 0.8–1.0 and then siFlex-C5 homogenizer anaerobically, and centrifuged at challenged with different stress for 30 min. Cultures were 13,200g for 20 min. Supernatant was transferred and harvested by centrifugation and stored in 20°C. Cell adjusted to 8 mL with degassed phosphate buffer. Cell-free pellets were resuspended and disrupted by sonication, the extract (2 mL) was mixed with 2 mL of 5% nitric acid con- supernatant containing soluble protein extract was then taining 0.1% (v/w) Triton X-100, followed by incubation quantified using the Bradford assay (Biorad, Hercules, and dilution as described above. The resulting sample con- CA, USA). For sodium dodecyl sulfate polyacrylamide gel tained both labile metal and protein-associated metal. The electrophoresis (SDS-PAGE), 50 lg of total protein was remaining 6 mL of supernatant was transferred to an loaded into each well and the gel was subjected to western Amicon Ultra-15 centrifugal tube with 3000 MW cutoff, blot analysis. Custom primary antibodies were generated and spun at 5000g. Two milliliter of flow-through was by Thermo scientific through rabbit immunization with applied to prepare the sample containing labile metal by heterologously expressed, purified B. subtilis SufC, SufB, following the same procedure as described above. and MnmA. The western blot membrane was incubated

ª 2015 The Authors. MicrobiologyOpen published by John Wiley & Sons Ltd. 619 Protective Role of Bacillithiol Z. Fang & P. C. Dos Santos with each primary antibody, secondary antibody followed et al. 2010; Fang et al. 2013). The last step of BSH bio- by Lumi-Phos WB chemiluminescent substrate, and then synthesis was suggested to be the condensation reaction exposed to Carestream Kodak film from Fisher to fix of L-Cys and N-glucosamine-malate (Helmann 2011). the protein bands. The films were then been scanned and Inactivation of bshA or bshC completely depleted intracel- images were processed by ImageJ software (Bethesda, lular BSH, whereas deletion of bshB only caused a modest MD, USA), each band area was quantified (Gassmann decrease in BSH levels. Initial studies showed that inacti- et al. 2009). The amount of protein in terms of lg was vation of BSH biosynthetic genes caused no effects on B. calculated based on a standard curve from pure proteins subtilis growth rates in LB-rich medium (Gaballa et al. in the range 0.02–0.1 lg. 2010). This observation suggested that the depletion of BSH did not cause major defects in metabolism. How- Results ever, it was also possible that the components in LB med- ium replaced the role of BSH and concealed phenotypes in mutant strains. To rule out this possibility, B. subtilis BSH null strains show a slow growth wild-type and mutant strains were grown in Spizizen’s phenotype in minimal medium minimal medium (MM). Interestingly, unlike the similar The biosynthesis of BSH was proposed to be a three-step growth profiles in LB (Fig. S1), the ΔbshA strain displayed reaction, utilizing BshA, BshB, and BshC enzymes. BshA a modest, yet-reproducible slower growth rate in MM as catalyzes the ligation of L-malate onto the carbon 1 posi- shown in Figure 1A. As bshA is located in an operon tion of UDP-N-acetyl-glucosamine to release UDP mole- along with ten other coding sequences including bshB, the cule (Parsonage et al. 2010). BshB is a metal-dependent slow growth phenotype could have been attributed to a deacetylase which cleaves the amide bond of N-acetyl-glu- polar effect on downstream genes. We have addressed this cosamine-malate to form N-glucosamine-malate (Gaballa potential concern in three experiments. First, the addition

(A) (B)

(C) (D)

Figure 1. Growth profile of Bacillus subtilis strain lacking BSH in MM. (A) Growth curves of wild-type B. subtilis (square) and ΔbshA strain (circle) in Spizizen’s MM (empty), MM plus 200 lmol/L of N-acetyl-glucosamine-malate (gray) or MM plus 200 lmol/L of BSSB (black). (B) Growth curves of wild type (square), ΔbshC amyE:PxylAbshC (triangle) in MM (empty), MM plus 2% xylose (black) and MM plus 200 lmol/L of BSSB (gray). (C)

Growth curves of wild type (square), ΔbshA (circle), and ΔbshC amyE:PxylAbshC (triangle) strains in MM (empty) and in MM plus 0.5% casamino acid (black). (D) Growth curves in the presence (black) and absence (empty) of 50 lmol/L of Fe2+ for wild type (square) and ΔbshA strain (circle). The curves shown are representatives of at least three independent experiments.

620 ª 2015 The Authors. MicrobiologyOpen published by John Wiley & Sons Ltd. Z. Fang & P. C. Dos Santos Protective Role of Bacillithiol of enzymatically produced N-acetyl-glucosamine-malate were comparable to the previously reported activities (GlcNAc-malate) (Fang et al. 2013), the product of the (Bohannon and Sonenshein 1989; Alen and Sonenshein BshA reaction, to cultures of ΔbshA strain recovered the 1999), and modestly reduced in the strain lacking BSH phenotype in MM (Figure 1A). Second, the growth rate (Fig. 2A). Bacillus subtilis genome encodes for one copy of a strain containing a deletion within another biosyn- of ACN which shows 51% sequence identity of the E. coli thetic gene, ΔbshC strain, showed a similar growth profile AcnA, which is the oxygen-resistant isoform of ACN. As in MM (Figure 1B). In B. subtilis, bshC is a monocystronic gene located remotely from other bshA and bshB biosyn- (A) thetic genes. Third, addition of BSSB, the oxidized form of chemically synthesized BSH, to either ΔbshA or ΔbshC cultures was able to repair these growth defects (Fig. 1A and B). Interestingly, addition of cystine or GSSG did not rescue the phenotype associated with lack of BSH. Collectively, these results confirmed that phenotypes associated with ΔbshA strain were attributed to the absence of BSH. Further studies revealed that the growth phenotype was fully restored with the supplement of 0.05% casamino acid (Fig. 1C). Via the screening of different combina- tions of amino acids, it was concluded that L-glutamine, L-glutamate and branched-chain amino acids L-leucine and L-isoleucine were able to correct the growth defect of (B) the BSH-deletion strain (Fig. S1). Interestingly, the addi- tion of 50 lmol/L Fe2+ also restored growth defects asso- ciated with lack of BSH in MM (Fig. 1D), whereas other divalent metals (Mn, Mg, Ca, and Zn) showed no drastic restoring effect (data not shown), confirming the specific- ity of Fe.

The depletion of BSH decreases the activities of Fe–S enzymes Initial growth experiments suggested the involvement of BSH in the synthesis of selected amino acids (leucine, iso- leucine, glutamine, and glutamate). The biosynthetic – pathways of these amino acids include Fe S enzymes. Figure 2. Lack of BSH leads to lower activity levels of Fe–S enzymes. Glutamine oxoglutarate aminotransferase (GOGAT) (A) Activities of the Fe–S proteins glutamate synthase (GOGAT), containing two 4Fe–4S clusters and one 3Fe–4S cluster aconitase (ACN), isopropyl malate isomerase (LeuCD) and the non-Fe– catalyzes the reversible conversion from L-glutamine and S protein malate dehydrogenase (MDH) in cell lysates of Bacillus D a-ketoglutarate to L-glutamate. LeuCD, a dihydroxy-acid subtilis wild type (black) and bshA strain (gray) cultured in MM (filled bar) or MM containing an additional 50 lmol/L of Fe2+ dehydratase in the biosynthesis of L-isoleucine, requires (diagonal striped bar). (B) Bacillus subtilis wild-type strain (black) and [4Fe–4S] cluster for its function. Likewise, the 4Fe–4S DbshA strain (gray) cultured MM and the activity of Fe–S enzymes enzyme ACN in TCA cycle also indirectly contributes to was quantified before (filled bar) and after the addition of 50 lmol/L synthesis of amino acids. The E. coli GSH null strain dis- of Fe2+ to cell lysates (upright striped bar). The activities of GOGAT, played a decrease in ACN activity (80% of WT levels) ACN, LeuCD, and MDH for wild-type B. subtilis were measured to be under standard growth conditions (Gardner and Frido- 40, 131, 17, and 1051 (nmol/min per mg total protein in crude vich 1993a). Therefore, we hypothesized that the growth extract), respectively. The activities of the DbshA strain were defects of the ΔbshA strain were caused by decreased Fe–S normalized to the activities of wild type in MM. All assays were performed in triplicates. The statistical analysis was performed using enzyme activities. To test this hypothesis, the specific unpaired t test, P values were obtained by comparing the DbshA activities of GOGAT, LeuCD, and ACN were determined without iron to wild-type w/o Fe and to DbshA with Fe. Comparisons in cell lysates from cultures harvested at mid-log phase of wild-type activities with and without Fe in both panels were not (OD600 of 0.8–1.0 when using 1 cm cuvette). The activ- statistical significant. (NS, not significant, *P < 0.05, **P < 0.01, ity levels of these enzymes in the wild-type cell lysates ***P < 0.001).

ª 2015 The Authors. MicrobiologyOpen published by John Wiley & Sons Ltd. 621 Protective Role of Bacillithiol Z. Fang & P. C. Dos Santos a control, the activity of MDH, a non-Fe/S enzyme, was for Fe–S enzymes (Macomber and Imlay 2009). As shown not significantly different in either strain. Supplementa- in Figure 3A, the addition of Cu(I) did not inhibit the tion of growth cultures with 50 lmol/L Fe2+, a condition initial growth rate of the wild-type strain, but decreased identified to suppress BSH phenotype in MM, led to an the OD600 in stationary phase from 0.4 to 0.265, display- increase in GOGAT, ACN, and LeuCD activities in ΔbshA ing similar growth inhibition as previously reported for B. strain by 28%, 17%, and 16%, respectively, whereas subtilis (Chillappagari et al. 2010). The BSH mutant, causing no effect to the levels detected for the wild-type however, showed a concentration-dependent inhibition of cell lysates (Fig. 2A). Similar results were also reported in the initial growth rate, which indicated the protective role Saccharomyces cerevisiae Δgsh1 strain, in which the growth of BSH against Cu(I) stress. The addition of Cu(I) slightly defect from deletion of GSH was also repaired upon addi- decreased the concentration of BSH (from 0.68 to tion of Fe3+ in the medium, and the LeuCD activity was 0.44 lmol/g DW) and BSSB (from 0.052 to 0.035 lmol/g also enhanced after 1-h exposure to Fe3+ (Kumar et al. DW), but it significantly stimulated protein S-bacillithio- 2011). In B. subtilis cultures, however, posthumous ex vivo lation by sixfold (from 0.02 to 0.12 lmol/g DW) (Fig. addition of 50 lmol/L of Fe2+ in the cell lysates did not 5A). Wild-type cells when cultured in MM and challenged improved the activity levels of GOGAT, ACN, and LeuCD with Cu(I) for 30 min showed reduction of 50% and in both the wild-type and ΔbshA strains (Fig. 2B). 87% of ACN and GOGAT activities (Fig. 4). This sup- In addition to Fe–S enzymes, mononuclear iron ported previous proposals that Fe–S enzymes were the enzymes are also known to be susceptible to changes in targets for Cu(I) stress in E. coli (Macomber and Imlay the cellular Fe status and found to respond to reactive 2009) and B. subtilis (Macomber and Imlay 2009; Chillap- oxygen species (Anjem and Imlay 2012). In this study, we pagari et al. 2010). However, in the mutant strain the determined the activity of two of these metalloenzymes, activities of these enzymes were not further affected with QD and Tdh. The lack of BSH caused a slight decrease in Cu(I) challenge (Fig. 4). their activities (Fig. S2). The addition of exogenous iron Cadmium ion (Cd(II)) has a high-binding affinity for into the cell lysates did not enhance the activities, which sulfur (Nies 2007). Therefore, the toxicity of Cd(II) has confirmed the preparation of cell lysate was strictly been suggested to be the result of the binding of Cd(II) anoxic. to free thiols, protein thiols, Fe–S centers and other sul- fur-rich compounds (Helbig et al. 2008). Cd(II) challenge affected the growth profile of both wild-type and mutant BSH participates in metal stress response strains, whereas displaying a more dramatic effect on the The effects of oxidative stress on the B. subtilis ΔbshA growth rate of the strain lacking BSH (Fig. 3B). Thiol strain have been previously investigated using zone inhi- analysis showed that despite the levels of BSH remaining bition assays demonstrating that both wild-type and at ~0.8 lmol/g DW within 50-min treatment, the BSSB BSH strain had the same sensitivity to H2O2 (Gaballa content increased sevenfold, from 0.07 lmol/g DW to et al. 2010). Herein, the role of BSH against a wide range 0.35 lmol/g DW (Fig. 5B). The activities of Fe–S enzymes of stresses was investigated in MM supplemented with were also determined in cell lysates of cultures subjected casamino acids. This growth medium not only ensured to 50 lmol/L Cd(II) challenge for 30 min. While the similar growth rates for both strains under no stress con- addition of Cd(II) to wild-type cultures decreased ACN dition (Figs. 1C, 3), but also eliminated the interference by only 20% (Fig. 4A), GOGAT was inactivated to nearly from potential protective biomolecules present in rich half of its initial activity (Fig. 4B). Similar to the effect medium. Initial screens involved growth assays in pres- observed for Cu(I) stress, the addition of Cd(II) did not ence of 0–300 lmol/L H2O2 (data not shown), which cause further damage to the activity of these Fe–S showed no differences between wild-type and DbshA enzymes in BSH null strain. Interestingly, deletion of BSH strains, in agreement with a previous report (Gaballa caused a 20% decrease in intracellular zinc which is a et al. 2010). Given the involvement of GSH in metal substrate for the CadA resistance system (Tsai et al. 1992; homeostasis, Fe–S biogenesis, and bacterial response to Ma et al. 2014). As a result, the CadA efflux ATPase was environmental stressors, we investigated the growth pro- repressed and more Cd(II) was accumulated inside cells, files of both B. subtilis wild-type and bshA deletion strains which may explain the observed elevated sensitivity of under conditions known to destabilize metal homeostasis BSH null strain to Cd(II). and Fe–S metabolism such as copper (Cu(I)), cadmium As exogenous addition of iron to cultures of B. subtilis (Cd(II)), as well Fe starvation conditions. DbshA recovered growth defects in MM and partially cor- Cu(I) can strongly trigger Fenton-like reactions to pro- rected biochemical defects associated with the loss of duce reactive hydroxyl radicals, which further cause a glo- BSH, we sought to determine the effects of iron starvation bal oxidative stress response, such as decreased activities in both strains. The growth profile of wild-type and

622 ª 2015 The Authors. MicrobiologyOpen published by John Wiley & Sons Ltd. Z. Fang & P. C. Dos Santos Protective Role of Bacillithiol

(A)

(B)

(C)

(D)

Figure 3. Growth curves of wild-type Bacillus subtilis (left panel) and ΔbshA strain (right panel) in MM with 0.05% casamino acid in presence of various stress challenges. (A) Cultures contained 0 lmol/L (square), 100 lmol/L (circle), and 300 lmol/L (triangle) of CuCl; (B) Cultures contained

0 lmol/L (square), 20 lmol/L (circle), and 50 lmol/L (triangle) of CdCl2. (C) Cultures contained 0 lmol/L (square), 100 lmol/L (circle), and 200 lmol/L (triangle) of 2,20-dypridyl. (D) Cultures contained 0 lmol/L (square), 100 lmol/L (circle), and 250 lmol/L (triangle) of paraquat. The curves shown are representatives of at least three independent experiments. mutant were examined under iron starvation conditions strain with 200 lmol/L DP (Fig. 3C). Nonetheless, DP in the presence of different concentrations of iron chela- challenge elicited response similar to what has been tor DP in MM supplemented with casamino acids. Only observed for Cu(I) and Cd(II) stresses (Fig. 4); that is, a minor inhibitory effect was observed in the mutant the addition of DP to growth medium resulted in 80%

ª 2015 The Authors. MicrobiologyOpen published by John Wiley & Sons Ltd. 623 Protective Role of Bacillithiol Z. Fang & P. C. Dos Santos

(A) by twofold (from 0.02 to 0.04 lmol/g DW) (Fig. 5C). PQ challenge has also been associated with iron starvation and NADPH depletion in B. anthracis (Pohl et al. 2011). Therefore, the total Fe content of cells cultured in the presence of PQ was determined. The mutant strain dis- played small decrease in the total Fe when compared with the wild type in the absence of PQ stressor (Fig. 6A). In both wild-type and DbshA strains, PQ challenge led to modest decrease in the total iron accompanied by minor increase in the Fe content associated to the LMW fraction (labile iron pool). In the presence of 100 lmol/L of PQ, the activity of GOGAT was more affected in the wild type than in the DbshA strain (Fig. 4B). The ACN activity, on (B) the other hand, remained unchanged in the wild-type strain, whereas displaying a nearly 50% decrease in cells extracts lacking BSH (Fig. 4A). Collectively, the PQ sensi- tivity associated with lack of BSH along with the thiol analysis of wild-type cultures exposed to PQ provides evi- dence for a protective role of BSH against superoxide stress.

BSH deficiency leads to Mn uptake under superoxide stress Oxidative stress has also been shown to trigger the import of Mn and induction of Mn-dependent superoxide dismutase. In E. coli, the expression of manganese super- Figure 4. Specific activities of ACN (A) and GOGAT (B) under stress oxide dismutase (MnSOD) is positively correlated with – conditions. Cells were grown in MM to late log phase (OD600 of 0.8 the redox ratio of GSH/GSSG (Gardner and Fridovich 1.0 measured with a 1 cm cuvette), at 37°C, and then 100 lmol/L of 0 1993b). The partial cross-regulation between thiol-redox 2,2 -dypridyl (DP), 100 lmol/L of paraquat (PQ), 100 lmol/L of H2O2 homeostasis and Fe/Mn levels (Helmann 2014), led us to (HP), 50 lmol/L of CdCl2 (Cd), or 100 lmol/L of CuCl (Cu) was added to challenge the cultures for 30 min. Cell lysates were determine whether the lack of BSH would affect cellular prepared as described in Experimental Procedures. The wild-type accumulation of these metals. Intracellular content of Fe strain is represented in black bars, whereas ΔbshA strain is shown in and Mn levels were determined in wild-type and DbshA gray bars. All assays were repeated in triplicates. The statistical cells cultured in MM (Fig. 6A and B). While the lack of analysis was performed using unpaired t test, P values compare BSH did not cause drastic changes in the levels of Fe ΔbshA sample with wild-type sample (NS, not significant, *P < 0.05, under stress conditions, PQ challenge elicited a time- **P < 0.01, ***P < 0.001). dependent increase in intracellular Mn content for DbshA cells. As shown in Figure 6B, the Mn level in the wild and 40% decrease in ACN and GOGAT activities in wild- type remained unaltered under PQ stress, whereas the type extracts, whereas causing modest to no inhibition in concentration of Mn in the mutant increased by more cells lacking BSH. than twofold over a 50-min treatment. This observation followed a similar response observed under oxidative challenge in E. coli, in which deletion of catalase and per- Bacillus subtilis strains lacking BSH are oxidase introduced oxidative stress that led to further sensitive to superoxide stress increase in intracellular Mn concentration (Anjem et al. Lastly, these strains were challenged with PQ, which can 2009). Based on phenotypic and biochemical analysis in induce superoxide stress in vivo. When treated with E. coli, it has been proposed that Mn could protect perox- 250 lmol/L of PQ, the mutant strain showed a remark- ide-stressed cells by replacing Fe in mononuclear enzymes able growth reduction, whereas wild type remained unaf- and by decreasing levels of Fe-triggered Fenton reaction fected (Fig. 3D). Thiol analysis also indicated that the (Anjem et al. 2009). In agreement with this model, PQ content of BSSB increased by near fivefold (from 0.07 to challenge led to increased levels of Mn associated with 0.34 lmol/g DW) and protein-associated BSH increased proteins, supporting a protection mechanism of Mn in B.

624 ª 2015 The Authors. MicrobiologyOpen published by John Wiley & Sons Ltd. Z. Fang & P. C. Dos Santos Protective Role of Bacillithiol

(A) (B)

(C)

Figure 5. Cellular BSH redox status upon metal and superoxide stress. Intracellular levels of BSH (square), BSSB (circle), and BSSProt (triangle) were determined in Bacillus subtilis wild-type cells cultured in MM before (time 0) and after challenge with 100 lmol/L of Cu (I) (A), 50 lmol/L of Cd (II) (B) or 100 lmol/L of paraquat (C). Oxidized BSH was determined when associated to intracellular small molecular weight thiol (BSSB) or large-molecule thiol (BSSProt) as described in the Experimental Procedures. The statistical analysis was performed using unpaired t test, P values compare time 0 sample with other samples (NS, not significant, *P < 0.05, **P < 0.01, ***P < 0.001). subtilis (Fig. 6C). In addition, the activity of MnSOD was BSH and Fe–S biosynthesis determined in cell lysates of wild-type and ΔbshA strains before and 30 min after 100 lmol/L PQ challenge. As In B. subtilis, the SUF system encoded by the sufCDSUB shown in Figure S3, the MnSOD activity in ΔbshA of operon, is proposed to be the only Fe–S cluster biosyn- 15.2 1.3 unit/min per mg was higher than activity thesis machinery (Albrecht et al. 2010; Selbach et al. detected for the wild-type strain (11.8 1.0 units/min 2010; Boyd et al. 2014). The lower levels of Fe–S clusters, per mg). This suggests that the deletion of BSH caused a judged by the lower activity of Fe–S enzymes, could also modest increase in MnSOD even under standard condi- be attributed to expression or inefficiency of biosynthetic tions (i.e., when cells were cultured in MM). PQ treat- apparatus in assembling Fe–S clusters in the absence of ment caused no significant change to MnSOD activity BSH and/or conditions detrimental to Fe–S metabolism. levels in wild type, but it led to a 1.5-fold increase in the Therefore, the levels of two biosynthetic components, strain lacking BSH (from 15.2 1.3 to 23.3 1.0 units/ SufC and SufB, were determined by western blot analysis min per mg). Activation of MnSOD elicited by PQ stress of lysates of cells cultured in MM. Expression levels of in the ΔbshA strain provided additional evidence for the both SufC and SufB in wild type were higher than the increase levels of protein-associated Mn determined mutant strain, whereas the internal control MnmA pro- through ICP-OES analysis. Besides the protective role of tein, proposed to participate in 2-thiouridine formation Mn mentioned above, early studies in B. subtilis suggested (Black and Dos Santos 2014), showed no significant that free Mn was also capable of eliminating the superox- change (Fig. S4). In E. coli, the SUF system is expressed ide stress through an unknown mechanism (Inaoka et al. only under oxidative stress or iron starvation to promote 1999). Therefore, ICP-OES analysis was also carried out Fe–S biogenesis under these conditions (Outten et al. for free Mn or that associated to LMW fraction Mn 2004). To verify the changes in expression levels of the B. (Fig. 6C). Levels of free Mn showed a similar increasing subtilis SUF components under stress conditions, western trend as the protein-associated Mn in response to PQ blot analyses of SufC were conducted in cell lysates of stress. These observations suggested a combined protec- MM cultures challenged with DP and PQ for 30 min. tion mechanism of Mn through metallating enzymes and Overall, SufC protein levels were similar across various increasing LMW Mn pool under PQ stress. conditions in the wild-type strain (Fig. 7). When the

ª 2015 The Authors. MicrobiologyOpen published by John Wiley & Sons Ltd. 625 Protective Role of Bacillithiol Z. Fang & P. C. Dos Santos

(A) (B)

(C)

Figure 6. Total iron (A) and manganese (B) concentration analysis in wild-type Bacillus subtilis (black bar) and ΔbshA strain (gray bar) after exposure to 100 lmol/L of paraquat stress. All measurements were performed in triplicate from independent cultures. The statistical analysis was performed using unpaired t test, P values compare ΔbshA sample with wild-type sample (NS, not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001). (C) Manganese distribution in ΔbshA strain under paraquat stress as shown labile (black bar) and protein- associated manganese (gray bar).

DbshA strain was challenged to various stresses, the levels conditions, did not show any significant difference (Gaba- of SufC increased to comparable levels to those observed lla et al. 2010). However, B. subtilis DbshA showed a slow in wild-type cultures (Fig. 7). This result suggested that growth phenotype in Spizizen’s MM, which was restored expression of sufC although constitutive can respond to upon addition of GlcNAc-malate and BSSB, supporting iron starvation and oxidative stress, but only when BSH the idea that growth defects observed in MM were attrib- is absent. uted to lack of BSH. Growth improvement was also observed upon supplementation with selected amino acids Discussion (Leu, Ile, Glu, and Gln) or iron. This initial result led us to explore a potential link between BSH and iron metabo- Since the discovery of BSH in Bacillus species in 2009, lism which includes Fe–S cluster biosynthesis. The results this biothiol has been proposed to be analogous to GSH showed that deletion of BSH caused severe growth defect and MSH in B. subtilis based on its similar structure and under both Cu(I) stress and PQ stress. As early as 1987, dominant concentration (Newton et al. 2009). Subsequent it has been reported that Fe–S cluster containing alpha, studies identified the involvement of BSH in protecting beta-dihydroxyisovalerate dehydratase in the isoleucine cells against hypochlorite, diamide, and fosfomycin chal- biosynthetic pathway was an target of superoxide (Kuo lenges (Gaballa et al. 2010; Chi et al. 2011, 2013; Posada et al. 1987). Subsequent studies also suggested that Fe–S et al. 2014). However, the stress-related protective mecha- dehydratases were primary targets of superoxide stress. nisms of BSH as well as other in vivo roles are still being Fe–S dehydratases can be oxidized at an extremely high unveiled. Initial growth assays with wild-type B. subtilis rate (3 9 106 mol L 1 s 1) (Flint et al. 1993), leading to and ΔbshA mutant strains in rich medium, under various intracellular superoxide overload and consequently a

626 ª 2015 The Authors. MicrobiologyOpen published by John Wiley & Sons Ltd. Z. Fang & P. C. Dos Santos Protective Role of Bacillithiol

encoding for the SUF components in B. subtilis are essen- tial for survival (Kobayashi et al. 2003; Albrecht et al. 2010). Therefore, we have investigated if expression of the housekeeping SUF components would also be subjected to regulation during iron starvation or oxidative stress. Unlike the expression profile observed for E. coli, SufC protein levels did not change significantly with oxidative challenge, suggesting that the B. subtilis SUF pathway was being constantly expressed. Our results are in agreement

Figure 7. Western blot analysis of Fe–S biosynthetic enzyme SufC with a recent transcriptomic study in B. subtilis cultured after challenge with dipyridyl (DP) and paraquat (PQ). Cultures were under 144 different conditions, which identified the suf < grown to OD600 of 0.8–1.0 and challenged with stressors for 30 min. genes among a small subset of coding sequences ( 3% of Cells were harvested, lysed. Clear extracts (50 lg of protein) were total genome) that were always expressed at levels higher subjected to SDS-PAGE and further analyzed by western blot. All than the chromosome median (Nicolas et al. 2012). While western blot assays were performed in triplicate of three independent this study indicated high levels of suf genes expression, it growth experiments. A representative image from western blot did not reveal conditions under which expression was fur- containing two separate gels of wild type and ΔbshA exposed in the same film is shown. ther augmented. Together, these observations provide additional support for the involvement of the B. subtilis branched-chain amino acid auxotrophy. The investigation SUF system in the essential housekeeping formation of of B. subtilis described in this study showed that similar Fe–S clusters, a role that is different from the specialized oxidative damage appeared to be exacerbated in the function of the E. coli SUF pathway that is only expressed absence of BSH, as evidenced by the PQ sensitivity of under stress conditions. The decreased activity of Fe–S the mutant strain. In addition to growth inhibition of the enzymes in cell lysates lacking BSH under MM growth DbshA strain, PQ challenge also resulted in increased lev- conditions indicates the potential involvement of BSH in els of intracellular Mn, disulfide pool and BSH-associated de the novo biosynthesis of Fe–S clusters. Unexpectedly, to proteins. In combination, these observations high- depletion of BSH caused a twofold reduction in SufC lighted the involvement of BSH in protecting cells against accumulation. It is possible that the lack of BSH induces superoxide stress. metabolic responses to protect the cell from oxidative Pathways containing Fe–S enzymes are susceptible tar- damage. It has been reported that bacterial responses to gets of oxidative stress. Bacterial responses to oxidative oxidative damage leads to down regulation of Fe- and stress includes (1) expression of hydrogen peroxide and Fe–S enzymes with concomitant expression of enzymes superoxide scavenging enzymes, (2) repression of Fe- and utilizing alternative cofactors. The increased levels of Mn- [Fe–S]-containing enzymes with concomitant expression SOD in the ΔbshA strain in the absence of any stressor of enzymes utilizing alternate cofactors, (3) mismetallation suggested that cells lacking BSH are responding to stress of enzyme’s Fe-binding site for Mn or Zn (Imlay 2006; even under standard growth conditions (e.g., minimal Helmann 2014), and (4) upregulation of Fe–S biosynthetic medium) and the lower expression of Fe–S biosynthetic pathways. In E. coli, the two Fe–S cluster biosynthetic components could be seen as another evidence of such machineries ISC and SUF use distinct regulatory strategies response. Interestingly, SufC expression levels in ΔbshA to respond to these environmental challenges (Outten strain raised to wild-type levels when cultures were chal- et al. 2004). The housekeeping ISC system supplies Fe–S lenged with an iron chelator. Bacillus subtilis lacks the clusters via its Fe–S cluster scaffold IscU. However, under Fe–S transcriptional regulator IscR (Giel et al. 2013; Mett- conditions of oxidative stress, transient clusters associated ert and Kiley 2014) and preliminary results described here to IscU are also prone to damage thus lowering the effi- suggest that this pathway may be subjected to distinct ciency of this system (Jang and Imlay 2010). Instead, yet-unidentified mechanism(s). under oxidative stress and Fe starvation conditions, Fe–S GSH has been proposed to be the major cytoplasmic cluster biogenesis is accomplished by the SUF system. In iron pool based on its binding affinity (Hider and Kong fact E. coli strains lacking SUF components are unable to 2011). In Salmonella enterica, GSH has been suggested to grow in the presence of DP and are sensitive to PQ and be a physiological chelator of the labile iron pool (Thor- hydrogen peroxide (Outten et al. 2004). gersen and Downs 2008), which could provide Fe for Fe– Bacillus subtilis and other gram-positive bacteria lack S cluster synthesis, export iron, present ferritin with the ISC pathway yet possess a modified SUF pathway excess iron, and supply iron directly to cytoplasmic which is thought to be the sole pathway for synthesis of enzymes (Andrews et al. 2003). In addition, GSH and Fe–S clusters in this group of bacteria. In fact, genes BSH have been shown to associate to proteins, causing

ª 2015 The Authors. MicrobiologyOpen published by John Wiley & Sons Ltd. 627 Protective Role of Bacillithiol Z. Fang & P. C. Dos Santos protection of cysteine residues from irreversible oxidation Bowater, L., S. A. Fairhurst, V. J. Just, and S. Bornemann. (Chi et al. 2011, 2013). In fact protein bacillithiolation 2004. Bacillus subtilis YxaG is a novel Fe-containing increases upon copper, cadmium, and PQ exposure. As quercetin 2,3-dioxygenase. FEBS Lett. 557:45–48. cysteine residues are typical ligands of metalloclusters, it Boyd, E. S., K. M. Thomas, Y. Y. Dai, J. M. Boyd, and F. W. is possible to consider that lack of BSH may impact Fe–S Outten. 2014. Interplay between oxygen and Fe-S cluster enzymes at a step prior cluster insertion. Furthermore, biogenesis: insights from the Suf pathway. Biochemistry the cysteinyl and malyl moieties of BSH are suitable metal 53:5834–5847. ligands (Newton et al. 2009). Although, it is hypothesized Bulaj, G., T. Kortemme, and D. P. Goldenberg. 1998. that BSH plays an analogous role to GSH in B. subtilis,it Ionization-reactivity relationships for cysteine thiols in is plausible to consider that BSH may also participate in polypeptides. Biochemistry 37:8965–8972. direct metal binding and mediate transport of Fe in bac- Chandrangsu, P., R. Dusi, C. J. Hamilton, and J. D. Helmann. terial species producing this biothiol. 2014. Methylglyoxal resistance in Bacillus subtilis: contributions of bacillithiol-dependent and independent pathways. Mol. Microbiol. 91:706–715. Acknowledgments Chesney, J. A., J. W. Eaton, and J. R. Mahoney. 1996. Bacterial The authors thank John Helmann and Ahmed Gaballa for glutathione: a sacrificial defense against chlorine – kind gifts of B.subtilis wild-type, DbshA, and DbshC compounds. J. Bacteriol. 178:2131 2135. strains and for providing with insightful comments to the Chi, B. K., K. Gronau, U. Mader, B. Hessling, D. Becher, and H. Antelmann. 2011. S-bacillithiolation protects against manuscript. This research was funded by the National hypochlorite stress in Bacillus subtilis as revealed by Science Foundation (MCB-1054623). P. D. S. and Z. F. transcriptomics and redox proteomics. Mol. Cell Proteomics have no conflict of interest to declare regarding commer- 10:M111 009506. cial or other relationships. Chi, B. K., A. A. Roberts, T. T. T. Huyen, K. Basell, D. Becher, D. Albrecht, et al. 2013. S-bacillithiolation protects Conflict of Interest conserved and essential proteins against hypochlorite stress in firmicutes bacteria. Antioxid. Redox Signal. 18: None declared. 1273–1295. Chillappagari, S., A. Seubert, H. Trip, O. P. Kuipers, M. A. References Marahiel, and M. Miethke. 2010. Copper stress affects iron homeostasis by destabilizing iron-sulfur cluster formation in Albrecht, A. G., D. J. A. Netz, M. Miethke, A. J. Pierik, O. Bacillus subtilis. J. Bacteriol. 192:2512–2524. Burghaus, F. Peuckert, et al. 2010. SufU is an essential iron- Cotgreave, I. A., R. Gerdes, I. Schuppe-Koistinen, and C. Lind. sulfur cluster scaffold protein in Bacillus subtilis. J. Bacteriol. 2002. S-glutathionylation of glyceraldehyde-3-phosphate 192:1643–1651. dehydrogenase: role of thiol oxidation and catalysis by Alen, C., and A. L. Sonenshein. 1999. Bacillus subtilis aconitase glutaredoxin. Methods Enzymol. 348:175–182. is an RNA-binding protein. Proc. Natl. Acad. Sci. USA Crack, J. C., C. D. den Hengst, P. Jakimowicz, S. 96:10412–10417. Subramanian, M. K. Johnson, M. J. Buttner, et al. 2009. Andrews, S. C., A. K. Robinson, and F. Rodriguez-Quinones. Characterization of [4Fe-4S]-containing and cluster-free 2003. Bacterial iron homeostasis. FEMS Microbiol. Rev. forms of Streptomyces WhiD. Biochemistry 48:12252–12264. 27:215–237. Crook, E. M. 1941. The system dehydroascorbic acid- Anjem, A., and J. A. Imlay. 2012. Mononuclear iron enzymes glutathione. Biochem. J. 35:226–236. are primary targets of hydrogen peroxide stress. J. Biol. Dalle-Donne, I., D. Giustarini, R. Rossi, R. Colombo, and A. Chem. 287:15544–15556. Milzani. 2003. Reversible S-glutathionylation of Cys 374 Anjem, A., S. Varghese, and J. A. Imlay. 2009. Manganese regulates actin filament formation by inducing structural import is a key element of the OxyR response to changes in the actin molecule. Free Radic. Biol. Med. 34: hydrogen peroxide in Escherichia coli.Mol.Microbiol. 23–32. 72:844–858. Dalle-Donne, I., R. Rossi, G. Colombo, D. Giustarini, and A. Black, K. A., and P. C. Dos Santos. 2015. Shared-intermediates Milzani. 2009. Protein S-glutathionylation: a regulatory in the biosynthesis of thio-cofactors: mechanism and device from bacteria to humans. Trends Biochem. Sci. functions of cysteine desulfurases and sulfur acceptors. 34:85–96. Biochim. Biophys. Acta 1853(6):1470–1480. Demasi, M., G. M. Piassa Filho, L. M. Castro, J. C. Ferreira, V. Bohannon, D. E., and A. L. Sonenshein. 1989. Positive Rioli, and E. S. Ferro. 2008. Oligomerization of the regulation of glutamate biosynthesis in Bacillus subtilis. cysteinyl-rich oligopeptidase EP24.15 is triggered by S- J. Bacteriol. 171:4718–4727. glutathionylation. Free Radic. Biol. Med. 44:1180–1190.

628 ª 2015 The Authors. MicrobiologyOpen published by John Wiley & Sons Ltd. Z. Fang & P. C. Dos Santos Protective Role of Bacillithiol

Fahey, R. C., W. C. Brown, W. B. Adams, and M. B. Imlay, J. A. 2006. Iron-sulphur clusters and the problem with Worsham. 1978. Occurrence of glutathione in bacteria. oxygen. Mol. Microbiol. 59:1073–1082. J. Bacteriol. 133:1126–1129. Inaoka, T., Y. Matsumura, and T. Tsuchido. 1999. SodA and Fang, Z., A. A. Roberts, K. Weidman, S. V. Sharma, A. manganese are essential for resistance to oxidative stress in Claiborne, C. J. Hamilton, et al. 2013. Cross-functionalities growing and sporulating cells of Bacillus subtilis. J. Bacteriol. of Bacillus deacetylases involved in bacillithiol biosynthesis 181:1939–1943. and bacillithiol-S-conjugate detoxification pathways. Jacob, C., I. Knight, and P. G. Winyard. 2006. Aspects of the Biochem. J. 454:239–247. biological redox chemistry of cysteine: from simple redox Flint, D. H., J. F. Tuminello, and M. H. Emptage. 1993. The responses to sophisticated signalling pathways. Biol. Chem. inactivation of Fe-S cluster containing hydro-lyases by 387:1385–1397. superoxide. J. Biol. Chem. 268:22369–22376. Jang, S. J., and J. A. Imlay. 2010. Hydrogen peroxide Fultz, P. N., and J. Kemper. 1981. Wild-type isopropylmalate inactivates the Escherichia coli Isc iron-sulphur assembly isomerase in Salmonella typhimurium is composed of two system, and OxyR induces the Suf system to compensate. different subunits. J. Bacteriol. 148:210–219. Mol. Microbiol. 78:1448–1467. Gaballa, A., G. L. Newton, H. Antelmann, D. Parsonage, H. Kennedy, M. C., M. H. Emptage, J. L. Dreyer, and H. Beinert. Upton, M. Rawat, et al. 2010. Biosynthesis and functions of 1983. The role of iron in the activation-inactivation of bacillithiol, a major low-molecular-weight thiol in Bacilli. aconitase. J. Biol. Chem. 258:11098–11105. Proc. Natl. Acad. Sci. USA 107:6482–6486. Kobayashi, K., S. D. Ehrlich, A. Albertini, G. Amati, K. K. Gardner, P. R., and I. Fridovich. 1992. Inactivation-reactivation Andersen, M. Arnaud, et al. 2003. Essential Bacillus subtilis of aconitase in Escherichia coli. A sensitive measure of genes. Proc. Natl. Acad. Sci. USA 100:4678–4683. superoxide radical. J. Biol. Chem. 267:8757–8763. Kumar, C., A. Igbaria, B. D’autreaux, A. G. Planson, C. Junot, Gardner, P. R., and I. Fridovich. 1993a. Effect of glutathione E. Godat, et al. 2011. Glutathione revisited: a vital function on aconitase in Escherichia coli. Arch. Biochem. Biophys. in iron metabolism and ancillary role in thiol-redox control. 301:98–102. EMBO J. 30:2044–2056. Gardner, P. R., and I. Fridovich. 1993b. NADPH inhibits Kuo, C. F., T. Mashino, and I. Fridovich. 1987. Alpha, beta- transcription of the Escherichia coli manganese superoxide- dihydroxyisovalerate dehydratase – a superoxide-sensitive dismutase gene (sodA) invitro. J. Biol. Chem. 268:12958– enzyme. J. Biol. Chem. 262:4724–4727. 12963. Lamers, A. P., M. E. Keithly, K. Kim, P. D. Cook, D. F. Stec, Gassmann, M., B. Grenacher, B. Rohde, and J. Vogel. 2009. K. M. Hines, et al. 2012. Synthesis of bacillithiol and the Quantifying Western blots: pitfalls of densitometry. catalytic selectivity of FosB-type fosfomycin resistance Electrophoresis 30:1845–1855. proteins. Org. Lett. 14:5207–5209. Giel, J. L., A. D. Nesbit, E. L. Mettert, A. S. Fleischhacker, B. Ma, Z., P. Chandrangsu, T. C. Helmann, A. Romsang, A. T. Wanta, and P. J. Kiley. 2013. Regulation of iron-sulphur Gaballa, and J. D. Helmann. 2014. Bacillithiol is a major cluster homeostasis through transcriptional control of the buffer of the labile zinc pool in Bacillus subtilis. Mol. Isc pathway by [2Fe-2S]-IscR in Escherichia coli. Mol. Microbiol. 94:756–770. Microbiol. 87:478–492. Macomber, L., and J. A. Imlay. 2009. The iron-sulfur clusters Giustarini, D., A. Milzani, G. Aldini, M. Carini, R. Rossi, and of dehydratases are primary intracellular targets of copper I. Dalle-Donne. 2005. S-nitrosation versus S- toxicity. Proc. Natl. Acad. Sci. USA 106:8344–8349. glutathionylation of protein sulfhydryl groups by S- Manikandan, K., A. Geerlof, A. V. Zozulya, D. I. Svergun, and nitrosoglutathione. Antioxid. Redox Signal. 7:930–939. M. S. Weiss. 2011. Structural studies on the enzyme Gu, M. Z., and J. A. Imlay. 2013. Superoxide poisons complex isopropylmalate isomerase (LeuCD) from mononuclear iron enzymes by causing mismetallation. Mol. Mycobacterium tuberculosis. Proteins 79:35–49. Microbiol. 89:123–134. Martinez-Ruiz, A., and S. Lamas. 2007. Signalling by Helbig, K., C. Grosse, and D. H. Nies. 2008. Cadmium toxicity NO-induced protein S-nitrosylation and S-glutathionylation: in glutathione mutants of Escherichia coli. J. Bacteriol. convergences and divergences. Cardiovasc. Res. 75: 190:5439–5454. 220–228. Helmann, J. D. 2011. Bacillithiol, a new player in bacterial Meister, A. 1988. Glutathione metabolism and its selective redox homeostasis. Antioxid. Redox Signal. 15:123–133. modification. J. Biol. Chem. 263:17205–17208. Helmann, J. D. 2014. Specificity of metal sensing: iron and Meister, A., and M. E. Anderson. 1983. Glutathione. Annu. manganese homeostasis in Bacillus subtilis. J. Biol. Chem. Rev. Biochem. 52:711–760. 289:28112–28120. Mettert, E. L., and P. J. Kiley. 2014. Coordinate regulation of Hider, R. C., and X. L. Kong. 2011. Glutathione: a key the Suf and Isc Fe-S cluster biogenesis pathways by IscR is component of the cytoplasmic labile iron pool. Biometals essential for viability of Escherichia coli. J. Bacteriol. 24:1179–1187. 196:4315–4323.

ª 2015 The Authors. MicrobiologyOpen published by John Wiley & Sons Ltd. 629 Protective Role of Bacillithiol Z. Fang & P. C. Dos Santos

Newton, G. L., K. Arnold, M. S. Price, C. Sherrill, S. B. Rey-Pailhade, J. D. 1888a. Nouvelle recherches physiologique Delcardayre, Y. Aharonowitz, et al. 1996. Distribution of sur la substance organique hydrogenant le soufre a froid. C. thiols in microorganisms: mycothiol is a major thiol in R. Acad. Sci. 107:43. most actinomycetes. J. Bacteriol. 178:1990–1995. Rey-Pailhade, J. D. 1888b. Sur un corps d’origine organique Newton, G. L., M. D. Unson, S. J. Anderberg, J. A. Aguilera, hydrogenant le soufre 1 a froid. C. R. Hebd. Seances Acad. N. N. Oh, S. B. delCardayre, et al. 1999. Characterization of Sci. 16:1683. Mycobacterium smegmatis mutants defective in 1-d-myo- Roberts, A. A., S. V. Sharma, A. W. Strankman, S. R. Duran, inosityl-2-amino-2-deoxy-alpha-d-glucopyranoside and M. Rawat, and C. J. Hamilton. 2013. Mechanistic studies of mycothiol biosynthesis. Biochem. Biophys. Res. Commun. FosB: a divalent-metal-dependent bacillithiol-S-transferase 255:239–244. that mediates fosfomycin resistance in . Newton, G. L., N. Buchmeier, and R. C. Fahey. 2008. Biochem. J. 451:69–79. Biosynthesis and functions of mycothiol, the unique Rodriguez-Pascual, F., M. Redondo-Horcajo, N. Magan- protective thiol of Actinobacteria. Microbiol. Mol. Biol. Rev. Marchal, D. Lagares, A. Martinez-Ruiz, H. Kleinert, et al. 72:471–494. 2008. Glyceraldehyde-3-phosphate dehydrogenase regulates Newton, G. L., M. Rawat, J. J. La Clair, V. K. Jothivasan, T. endothelin-1 expression by a novel, redox-sensitive Budiarto, C. J. Hamilton, et al. 2009. Bacillithiol is an mechanism involving mRNA stability. Mol. Cell. Biol. antioxidant thiol produced in Bacilli. Nat. Chem. Biol. 28:7139–7155. 5:625–627. Roos, G., N. Foloppe, and J. Messens. 2013. Understanding the Nicolas, P., U. Mader, E. Dervyn, T. Rochat, A. Leduc, N. pK(a) of redox cysteines: the key role of hydrogen bonding. Pigeonneau, et al. 2012. Condition-dependent transcriptome Antioxid. Redox Signal. 18:94–127. reveals high-level regulatory architecture in Bacillus subtilis. Selbach, B., E. Earles, and P. C. Dos Santos. 2010. Kinetic Science 335:1103–1106. analysis of the bisubstrate cysteine desulfurase SufS from Nies, D. H. 2007. Bacterial transition metal homeostasis. Pp. Bacillus subtilis. Biochemistry 49:8794–8802. 118–142 in Molecular microbiology of heavy metals. Sen, C. K., and L. Packer. 2000. Thiol homeostasis and D.H.N.a. and S. Silver, ed. Springer Berlin Heidelberg, supplements in physical exercise. Am. J. Clin. Nutr. Germany. 72:653S–669S. Outten, F. W., O. Djaman, and G. Storz. 2004. A suf operon Sharma, S. V., V. K. Jothivasan, G. L. Newton, H. Upton, J. I. requirement for Fe-S cluster assembly during iron starvation Wakabayashi, M. G. Kane, et al. 2011. Chemical and in Escherichia coli. Mol. Microbiol. 52:861–872. chemoenzymatic syntheses of bacillithiol: a unique low- Pan, S., and B. C. Berk. 2007. Glutathiolation regulates tumor molecular-weight thiol amongst low G + C Gram-positive necrosis factor-alpha-induced caspase-3 cleavage and bacteria. Angew. Chem. Int. Ed. Engl. 50:7101–7104. apoptosis: key role for glutaredoxin in the death pathway. Sharma, S. V., M. Arbach, A. A. Roberts, C. J. Macdonald, M. Circ. Res. 100:213–219. Groom, and C. J. Hamilton. 2013. Biophysical features of Parsonage, D., G. L. Newton, R. C. Holder, B. D. Wallace, C. bacillithiol, the glutathione surrogate of Bacillus subtilis and Paige, C. J. Hamilton, et al. 2010. Characterization of the other firmicutes. Chembiochem 14:2160–2168. N-acetyl-alpha-D-glucosaminyl l-malate synthase and Siegel, S. E. a. L. 1969. Mitochondrial L-malate dehydrogenase deacetylase functions for bacillithiol biosynthesis in Bacillus of beef heart. Methods Enzymol. 18:99–106. anthracis. Biochemistry 49:8398–8414. Thorgersen, M. P., and D. M. Downs. 2008. Analysis of yggX Pohl, S., W. Y. Tu, P. D. Aldridge, C. Gillespie, H. Hahne, U. and gshA mutants provides insights into the labile iron pool Mader, et al. 2011. Combined proteomic and transcriptomic in Salmonella enterica. J. Bacteriol. 190:7608–7613. analysis of the response of Bacillus anthracis to oxidative Tsai, K. J., K. P. Yoon, and A. R. Lynn. 1992. ATP-dependent stress. Proteomics 11:3036–3055. cadmium transport by the cadA cadmium resistance Posada, A. C., S. L. Kolar, R. G. Dusi, P. Francois, A. A. determinant in everted membrane-vesicles of Bacillus Roberts, C. J. Hamilton, et al. 2014. Importance of Subtilis. J. Bacteriol. 174:116–121. bacillithiol in the oxidative stress response of Staphylococcus Winterbourn, C. C., and D. Metodiewa. 1999. Reactivity of aureus. Infect. Immun. 82:316–332. biologically important thiol compounds with superoxide Rawat, M., G. L. Newton, M. Ko, G. J. Martinez, R. C. Fahey, and hydrogen peroxide. Free Radic. Biol. Med. 27:322–328. and Y. Av-Gay. 2002. Mycothiol-deficient Mycobacterium smegmatis mutants are hypersensitive to alkylating agents, Supporting Information free radicals, and antibiotics. Antimicrob. Agents Chemother. 46:3348–3355. Additional Supporting Information may be found in the Reed, D. J. 1990. Review of the current status of calcium and online version of this article: thiols in cellular injury. Chem. Res. Toxicol. 3:495–502.

630 ª 2015 The Authors. MicrobiologyOpen published by John Wiley & Sons Ltd. Z. Fang & P. C. Dos Santos Protective Role of Bacillithiol

Figure S1. Growth profiles of Bacillus subtilis wild type was detected through formation of formazan dye using (square), ΔbshA (circle), and ΔbshC AmyE:Pxyl-bshC (tri- the superoxide dismutase kit from Cayman Chemical, fol- angle) strains in LB (black) and Spizizen’s MM plus Leu, lowing the manufacture’s protocol. One unit is defined as Ile, Gln, and Glu (empty). The growth curves shown are the amount of enzyme needed to exhibit 50% dismuta- representative of at least of three independent experi- tion of the superoxide radical. Statistical analysis was per- ments. formed using unpaired t test, P values were obtained by Figure S2. Activities of mononuclear iron enzymes threo- comparing ΔbshA sample with wild-type sample (NS, not nine dehydrogenase (A) and quercetin 2,3-dioxygenase significant, **P < 0.01). (B). Activities were determined with wild-type Bacillus Figure S4. Expression of SufB, SufC and MnmA in cell subtilis (black) and ΔbshA strain (gray) cell lysates from lysates of wild-type and ΔbshA strains. Cell lysates were MM (filled bar) and MM plus 50 µmol/L of exogenous prepared as described in Experimental Procedures. Each Fe2+ (upright striped bar). All assays were performed in assay was performed in triplicate. For each wild-type and triplicates. The statistical analysis was performed using ΔbshA cell lysate, 50 µg of protein were loaded into the unpaired t test, P values were obtained by comparing same gel. The expressions of SufC, SufB, and MnmA were each sample with the ΔbshA without Fe. Comparison of determined from the same cultures. Each blot containing wild-type cultures with and without Fe was not statically wild-type and ΔbshA cell lysates and known concentra- significant. NS, not significant, *P < 0.05, **P < 0.01, tions of each respective purified protein were run in the ***P < 0.001, ****P < 0.0001. same gel and exposed in the same film. A presentative Figure S3. MnSOD activity in the cell lysates of CU1065 western blot is shown above. The bands were quantified (wt) and HB11002 (bshA) before (black) and 30 min through ImageJ and the statistical analysis was performed after (gray) the treatment of 100 µmol/L paraquat. The using unpaired t test, P values were obtained by compar- cultures were grown to OD600 of 1.0 in MM, and samples ing ΔbshA sample with wild-type sample (NS, not signifi- were taken before and 30 min after the treatment of cant, *P < 0.05, **P < 0.01, ***P < 0.001, 100 µmol/L paraquat. Cell pellets were resuspended, ****P < 0.0001). lysed, and assayed for superoxide dismutase. The activity

ª 2015 The Authors. MicrobiologyOpen published by John Wiley & Sons Ltd. 631